Av

BF

a

ARRA

KBME

1

cudtt[

frdHia

ipaIt

0h

Energy and Buildings 55 (2012) 618–630

Contents lists available at SciVerse ScienceDirect

Energy and Buildings

j our na l ho me p age: www.elsev ier .com/ locate /enbui ld

numerical modeling approach to evaluate energy-efficient mechanicalentilation strategies

ilal Yassine, Kamel Ghali ∗, Nesreen Ghaddar, Issam Srour, Ghassan Chehabaculty of Engineering and Architecture, American University of Beirut, P.O. Box 11-0236, Beirut 1107-2020, Lebanon

r t i c l e i n f o

rticle history:eceived 30 June 2012eceived in revised form 9 August 2012ccepted 28 August 2012

eywords:uilding materialsechanical ventilation

a b s t r a c t

This paper investigates design features that can potentially reduce the energy consumed in attainingappropriate thermal comfort levels in typical residential buildings. A numerical model coupled with aPID controller is developed to predict the indoor air temperature that is adjusted via mechanical ventila-tion. The model is used to simulate and evaluate various scenarios of building wall layouts and materials.From the various simulation runs, the wall configurations and materials were refined to combinationsthat rendered the mechanical ventilation a feasible option for attainment of comfort for the largest num-ber of hours per year. The runs were conducted for a typical residential apartment located in the city

nergy efficiency of Beirut, Lebanon, which is characterized by mild coastal climatic conditions. Different wall configura-tions were assumed for each of the living zone and the bedroom zone of the apartment. The simulationresults suggest an optimal wall configuration comprised of a 5 cm layer of insulating strawboard sand-wiched between a 2 cm × 10 cm wall made of masonry units consisting of Hempcrete (mixture of Portlandcement, aggregates, and industrial hemp fibers) for the living zone, whereas the wall for the bedroomzone consists of a 10 cm of Hempcrete.

. Introduction

Worldwide, the building sector accounts for 50% of total energyonsumption [1]. This percentage is expected to increase with pop-lation growth and climate change compounding the problem ofwindling energy resources. In an effort to minimize this increase inhe building’s energy demand, there is a growing movement towardhe design, construction, and operation of energy efficient buildings2–4].

The majority of the research effort in this field [5–13] hasocused on methodologies, innovations, and initiatives aimed ateducing the operational energy consumption. Examples includeevelopment of energy efficient electro-mechanical systems (e.g.VAC units, lighting), use of renewable energy, and adoption of

ncentives for use of building envelopes that provide high insulationnd air tightness.

One means of mitigating the increase in energy consumptions by relying on mechanical ventilation systems for cooling pur-oses and on the applications of natural (as compared to processed

nd manufactured) and recycled materials in building construction.ncorporation of natural ventilation systems and ceiling fans leado a decrease in the usage of mechanical air conditioning systems

∗ Corresponding author. Tel.: +961 1 350000x3438; fax: +961 1 744462.E-mail address: ka04@aub.edu.lb (K. Ghali).

378-7788/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.enbuild.2012.08.042

© 2012 Elsevier B.V. All rights reserved.

and help attain indoor air quality and thermal comfort at higher airtemperatures.

Several studies have shown that mechanical ventilation con-tributes to energy efficient buildings especially for buildings withappropriate thermal mass and sufficient daily temperature swingfrom day to night. On the other hand, the use of natural mate-rial such as industrial hemp fibers mixed with concrete has beenreported to have low heat conductivity in comparison with otherbuilding materials in construction; thus, it reduces the summerheat gains and winter heat losses [5].

Several studies have investigated the impact of ventilation onthe internal indoor air temperature [6–9]. The key parametersrelated to the efficiency of ventilation can be divided into threecategories: climatic factors, building material and control strat-egy parameters. An example of climatic factors is the outdoor airtemperature, which includes mean daily temperature and diur-nal temperature range, i.e. the climatic potential index. Buildingparameters include the building type, envelope openings such aswindows and doors, building material and building structure. Theflow rate and time period are determined by the control strategywhen mechanical ventilation is applied (night ventilation, day timeventilation, full-day ventilation and no ventilation).

Kolokotroni and Aronis [6], Blondeau et al. [7], Breesch [8], Car-rilho da Graca et al. [9] and Eicker [10] concluded that the use ofnight ventilation contributes to energy efficient buildings espe-cially for buildings with appropriate thermal mass and sufficient

B. Yassine et al. / Energy and Bui

Nomenclature

A area of the wall (m2)Cp specific heat (J/kg K)COP coefficient of performanceG rate of moisture generation inside the spaceh heat convection coefficient (W/m2 K)I total solar radiationK thermal conductivity (W/m K)Kc proportional gain parametersm mass flow ratePID proportional–integral–derivative controllerT temperature in (K)TI integrative gain parameterTD derivative gain parameterT time (s)qint internal heat loadV volume of the space (m3)W humidity ratio (kg H2O/kg dry air)X axial direction

Greek letters� density

wall surface absorptivity

Subscriptsa airW wall

dtawtt[mtwa

atmocc(o(iniw

ectttmca

∞ ambient

aily temperature swing from day to night. Similarly, daytime ven-ilation (mechanical and natural) was determined to be applicablend energy efficient for all buildings. However, in regions that lackind speed such as high density cities where the building’s struc-

ure blocks the wind field, replenishing indoor air for occupanthermal comfort can be achieved only by mechanical ventilation11–13]. Zhou et al. [14] discussed the effect of coupling the thermal

ass and natural ventilation on the passive building design. Theyested different wall configurations and found that heavy wallsith external insulation provide the lowest amplitude of indoor

ir temperature among different external low mass walls.In terms of natural material, wood, sisal, jute, bamboo, coconut,

sbestos, rockwool and hemp are examples that can be used. Alonghose lines, an innovative building material that has shown to yield

ultiple socio-economic benefits is Hempcrete. Hempcrete is a mixf cement mortar with fibers of industrial hemp, a sister plant ofannabis. The benefits of using hemp have been reported by a studyonducted by Awwad et al. [15]. The study shows that Hempcrete:i) improves physical characteristics and structural performancef concrete; (ii) reduces raw materials and energy resources andiii) provides a material with better thermal property and thereforencreases energy efficiency. With such benefits, and being a carboneutral material, the use of hemp in concrete reduces the embod-

ed energy of concrete and other cementitous materials which areidely used and in large quantities in building construction.

This paper further investigates options to lower operationalnergy without negatively affecting embodied the energy of typi-al apartments in urban and rural settings by conducting researchhat considers both the building materials and a mechanical ven-ilation system for comfort and air quality. The paper focuses,

hrough a case study in Lebanon on the use of local and regional

aterials such as strawboard for insulation and hemp in con-rete that contribute to the decrease in the building embodiednd operational energy. In addition, the paper proposes a space

ldings 55 (2012) 618–630 619

model for controlling the amount of ventilation air needed for con-ditioning the building’s indoor environment. The paper does notconsider natural ventilation as there are many uncontrollable vari-ables (blockage of urban wind by high rise building, uncontrolledoutside air velocity, security issues) which could affect its reliabil-ity as an effective air conditioning option. Mechanical ventilationand ceiling fans are used to adjust the indoor air temperature andspeed whenever needed to achieve thermal comfort. Any fluctua-tions in outside temperature or solar load can cause the internalair temperature to fluctuate in a similar way, though delayed anddampened depending on the thermo-physical properties (thermalcapacitance, resistance and density) of the building wall construc-tion materials as well as its layering arrangement. The effectivenessof mechanical ventilation in achieving thermal comfort for a givenbuilding construction material is determined through the reliabil-ity of this system in offering thermal comfort without having theneed to resort to mechanical cooling during summer and winteroperations.

2. Mathematical formulation

To meet the objective of this paper, a mechanical ventilationcontroller integrated with a numerical multi-zonal space and com-fort model is developed (Fig. 1). For given building materials anda set of indoor and outdoor conditions, the ventilation controllermodulates the amount of ventilation air needed to moderate theindoor air temperature and minimize discomfort hours. Differ-ent building materials are assessed for both their effect on indoorcomfort and fan consumption. The numerical space model is vali-dated against other commercial simulation software that simulatesindoor air temperature for Lebanon conditions.

2.1. Space model

The space analyzed in the case study is divided into two zones:living and bedroom. The transient one-dimensional heat conduc-tion equation for a multilayered wall consisting of N parallel layersis given by the following equation:

�i,jci,j∂Ti,j(t, x)

∂t= ki,j

∂2Ti,j(t, x)

∂x2(1)

where i represents building element (wall, ceiling, or floor), j rep-resents the jth layer within the element i, t and x are the time andspatial coordinates respectively, Tij is the temperature, whereas k,�, and c are the thermal conductivity, density, and thermal capacityrespectively.

The boundary conditions for the outdoor and indoor surfaces ofthe wall are

· I(t)+hc∞[T∞(t)−Ti(0, t)] + hr∞[Tsky(t) − Ti(0, t)] = −k∂Ti,j(0, t)

∂x(2)

hc a[Ti(L, t) − Ta(t)] +6∑

j=1

hr i−j[Ti(L, t) − Tj(t)] = −k∂Ti(L, t)

∂x(3)

where the external radiative coefficient, hr∞, is equal to ε�(Tsky +Ti(0, t))(T2

sky+ T2

i(0, t)) and the internal radiative coefficient, hr i−j ,

is equal to ε�Fi−j(Ti(L, t) + Tj(t))(T2i

(L, t) + T2j

(t)). The coefficientshca and hr are the convective and radiative heat transfer coefficientsfor the outdoor and indoor wall surfaces respectively, is the wall

surface absorptivity, and I is the total solar radiation on the wallside. The temperatures T∞, T(0, t), Ta(t), T(L, t) and Tsky are the ambi-ent air and outside wall surface temperature, room air, inner wallsurface temperature, and sky temperature.

620 B. Yassine et al. / Energy and Buildings 55 (2012) 618–630

syste

a

�

wotaq

f

�

wlr

2

i5sposdtitsdamaaCt

Fig. 1. Schematic of the ventilation

Neglecting infiltration, the internal lumped air node energy bal-nce is given by

aVacp,a∂Ta(t)

∂t=

6∑i=1

hc,a · Ai(Ti − Ta) + mcp,a(T∞ − Ta) +∑

qint

(4)

here �a, Va, and ca are density, volume and specific heat capacityf the room air respectively, Ai and Ti are the surface area and theemperature of the room element i, m and T∞ are the mass flow ratend the temperature of the air supplied by the ventilation system.int is the internal heat load.

Eq. (5) represents simplified dynamic moisture for the space, asollows:

aVad wa

∂t= G − m(wa − w∞) (5)

here G is the rate of moisture generation inside the space due toatent loads, wa and w∞ are the room air and supply air humidityatios.

.2. Thermal comfort model

Different thermal comfort model are discussed in several stud-es [16–20]. The new adaptive comfort model for ASHRAE standard5 address thermal comfort in naturally ventilated building. In thistandard the temperature range that satisfies thermal comfort ofeople is higher than the other conservative approaches basedn the predicted mean vote (PMV) method [18,19]. However, thistandard is limited to building with reachable and operable win-ows that is controlled by human. For this reason, in this researchhe predicted mean vote (PMV) model [18] is used to assess thendoor thermal comfort. The PMV represents the thermal sensa-ion of people on a scale from −3 (cold) to +3 (hot). The model isimple, widely used by researchers and its input requirements areirect and can be categorized as environmental (temperature, rel-tive humidity, air velocity,) and personal (clothing insulation andetabolic rate). Thermal comfort inside the space is calculated on

n hourly basis, based on the room hourly average temperature,verage relative humidity and air velocity for the precedent hour.omfort inside the space is assumed to occur when the PMV insidehe space is between −1 and +1.

m coupled with the PID controller.

2.3. Controller equation

A PID controller is used to modulate the ventilation air flow ratebased on room temperature and indoor fresh air requirement [21].Every 5 min, the controller updates the ventilation air flow rate tokeep the indoor air temperature at its summer and winter set pointsand to provide thermal comfort for longer periods, if possible, asindoor air temperature deviates from its set point.

The indoor air temperature is controlled using the followingequation:

m(t) = m(t−1) +(

Kc

(1 + �t

TI+ TD

�t

)× (Troom − Tset point)

−(

1 + 2TD

�t

)× (Troom − Tset point)(t−1) + TD

�t

×(Troom − Tset point)(t−1)

)(6)

where �t is the time step, Kc, TI and TD are the proportional, integra-tive and derivative gains parameters. These parameters are tunedto track the setpoint temperatures by determining the amount ofair flow rate needed to minimize the difference between the indoorair temperature and the setpoint temperature.

3. Numerical simulation methodology

As mentioned earlier, the space and comfort models are inte-grated with the ventilation controller model into one transientmodel. The model is capable of determining the amount of ventila-tion air that can possibly provide indoor thermal comfort for longerperiods and prevent large deviations of indoor air temperature fromits set point during summer and winter operations. The transientmodel is simulated to obtain the indoor zone thermal comfort whensubjected to variable external and internal loads. This requires thefollowing information: the space dimensions, characteristics of theconstruction materials, outdoor weather conditions, external andinternal loads and the ventilation controller operation conditions

(e.g. activation and deactivation modes). Eq. (1) is solved numeri-cally using the finite difference method with a fully implicit schemewhere each wall layer is decomposed into a number of volumes,each having one node. The program is developed using MATLAB©

d Buildings 55 (2012) 618–630 621

(n

ocrteldistea

3

rivimwShlswop(

8ocfltlmnmbrii

tfittT(tvotttoit

a

B. Yassine et al. / Energy an

2010). The total number of nodes used in each wall layer is 120odes.

Starting from arbitrary initial conditions, and using a time stepf 2.5 min, the inner surface temperatures of the space walls areomputed along with the inner air temperature. This time stepepresents a compromise between accuracy and simulation run-ime. Simulations are performed for typical weather conditions ofach month to obtain the effectiveness of the mechanical venti-ation system in providing thermal comfort throughout the entireay. At the end of the 24 h operational period of each month, the

nitial conditions are recalculated and used as input in the cyclicimulations until steady periodic convergence is achieved. The cri-erion for convergence is reached when the maximum percentagerror for air temperature between the simulation values at time tnd t + 24 h is less than 10−4%.

.1. Control strategy

The strategy that is used to condition the indoor environmentelies on controlling the amount of air brought to the space dur-ng summer and winter operation. The outdoor air is allowed toary between a minimum and an upper limit. The minimum limits determined by the fresh air requirement of 7.5 l/s/person (recom-

ended by ASHRAE [16]), whereas the upper limit is set to 25 ACHhich is close to what has been reported by Geros et al. [11] and

haviv et al. [12] as the maximum effective rate of air change for anour. The controller varies the amount of fresh air between the two

imits to maintain an indoor temperature close to the winter andummer set points whenever the outdoor conditions allow. Other-ise, the controller prevents large deviations between indoor and

utdoor temperatures. The control strategy depends on differentarameters such as: the type of zone (living or bedroom), the timeday or night) and the season (winter or summer).

For the living zone, the control is active during the day (after in the morning) and remains active till hour 23. During non-ccupancy period (23–8), in the summer season, night purging withonstant air flow is applied into the space; the supplied outdoor airow takes advantage of the low temperature profile during nightime throughout the summer season and helps purging the internaloads accumulated in the space during the day, while in winter the

inimal requirement are applied to the space. Different values foright purging have been used in literatures [6,10,12]. During sum-er season, an air change per hour of 5 ACH, is close to what has

een used in the Middle East region [12] and close to what has beeneported by Kolokotroni and Aronis [6] and Blondeau et al. [7], ands therefore used in this paper. As for the bedroom zone, the controls active during the occupancy time (23 till 10).

The season is a main parameter in the controller strategy; thus,he following strategy is used during the activation time period:or the winter period (November through March), the controllers set to activate whenever the indoor air temperature is lowerhan the outdoor ambient temperature, provided that it is lowerhan the winter set point temperature of 22 ◦C, (Troom < Tamb androom < Tset point). On the other hand, during the summer periodApril through October), if the indoor air temperature is lower thanhe outdoor ambient temperature (Troom < Tamb), the amount ofentilation air entering the space is set to the minimum amountf fresh air requirement recommended by ASHRAE [16]. The con-roller is activated whenever the indoor air temperature is higherhan the outdoor ambient temperature (Troom > Tamb) provided thathe indoor air temperature is higher than the set point temperaturef 26 ◦C (Troom > Tset point). If thermal comfort is not attained, a ceil-

ng fan is turned on concurrently to increase the indoor air velocityo 1.5 m/s.

In order to minimize the fluctuation resulting from consecutivectivation and deactivation of the controller, the air temperature

Fig. 2. Numerical model flow chart (urban setting).

and the mass flow rate at time (t) are compared to those at time(t − 1). If the difference is less than 5% for one of the two parameters,then the same mass flow rate at time (t − 1) is used at time (t) andthe indoor air temperature is simulated for this amount of air flow.Fig. 2 presents the flow chart of the proposed numerical model.

4. Validation of the numerical model

To ensure that the results of the numerical model are realisticand represent the real conditions of a typical residential apart-ment in Beirut, Lebanon, the numerical model should be validatedagainst calibrated software such as TRNSYS which is known for itsflexibility in modeling mechanical ventilation [22]. This requires

making sure that the results of the numerical model are accurateand also simulate the actual conditions of residential apartmentsin Beirut. Therefore, TRNSYS was first calibrated and then used tovalidate the numerical space model [22]. To calibrate TRNSYS, a

622 B. Yassine et al. / Energy and Buildings 55 (2012) 618–630

artme

tat[ioc

ttriTTtT

Fig. 3. Typical ap

ypical apartment in Beirut (base case) is simulated, and the resultsre compared against published data on energy consumption for aypical residential apartment in Lebanon city by Houri and Orfali23] and Karaki et al. [24]. This is done by entering the data of a typ-cal residential house in Beirut, including building structural data,perational and occupancy schedule, internal loads and outdoorlimatic conditions.

Once calibrated, results from TRNSYS can serve as a benchmarko validate the results obtained from the numerical model, in awo-step process. The first step is to run the simulation software,esulting in 181 values (180 values during occupancy period (15 h)n addition to a constant flow rate during non-occupancy (8 h)).

hese values are then input from the numerical model into TRNSYS.he second step is to compare the simulated indoor air tempera-ure provided by the numerical model to that provided by TRNSYS.he following sections elaborate on the validation effort.

Fig. 4. Schedule of the people lo

nt plan in Beirut.

4.1. Base case description

As mentioned earlier, a typical apartment in Beirut is consid-ered as a base case in this paper. The apartment is assumed to havea 16 m × 10 m square layout and a height of 3 m as shown in Fig. 3.It is divided into two zones: living zone (dining and living rooms)and bedroom zone (three bedrooms). The living zone has an areaof 100 m2 while the bedroom zone has an area of 60 m2. The zoneshave a common partition wall and 3 external walls each. The exter-nal walls are made of 15 cm concrete masonry blocks with 1.5 cmlay mortar cement plaster on the internal and external sides of thewall, while the internal partition wall is only 10 cm thick. The exter-

2

nal walls have an overall heat transfer coefficient of 2.36 W/m Kand a heat capacity of 1200 J/kg and the internal partitionhas anoverall heat transfer coefficient of 3.63 W/m2 K, which is typical forurban residential apartments in Lebanon [25,26]. The living zone

ad inside the apartment.

B. Yassine et al. / Energy and Buildings 55 (2012) 618–630 623

trical

hdnw

usTstz

4

cal1TrE

mstp

Fig. 5. Schedule of the elec

as three external walls oriented to the North, South and Westirection with an area of 30 m2, while the bedroom has two exter-al walls of 18 m2 area oriented the North and South, and a 30 m2

all oriented to the East.The internal load component is based on the occupancy sched-

le of a typical Lebanese family consisting of 6 persons [25]. Fig. 4hows the hourly variation in occupancy between the two zones.he electrical loads are 10 W/m2 of sensible load for lighting fixtureand 800 W for appliances (refrigerator, TV and PC). Fig. 5 presentshe schedule of the electrical load in both the living and bedroomones. The schedule is assumed to apply for the whole year.

.2. Base case calibration

A simulation of the base case using TRNSYS [22] showed a peakooling load of 14.05 kW occurring at hour 16 on July 28th, with annnual cooling requirement of 132.8 kWh/m2, and a heating peakoad of 5.49 kW occurring at hour 6 in the morning on February2th with an annual heating requirement equal to 27.8 kWh/m2.he weather data (outdoor air temperature, solar radiation and theelative humidity) is obtained from the records of the Mechanicalngineering Department at the American university of Beirut.

The simulations are made on an intermittent basis. During sum-

er season, the cooling mode is activated and the heating mode is

witched off; however, the system is scheduled to turn off whenhe temperature inside the space is lower than the set point tem-erature (26 ◦C). On the other hand, during the winter season, the

Fig. 6. Reported and simulated monthly ele

load inside the apartment.

heating mode is activated and the cooling mode is switched offwhen the temperature is higher than the set point temperature(22 ◦C). The total base case heating and cooling needs during thewhole year are equal to 160.60 kWh/m2/year of thermal energy.Using a typical energy performance of a split unit HVAC system atan average COP of 3.0, the total annual electric energy consump-tion of the cooling and heating system in the base case is equal to53.53 kWh/m2.

The total electrical energy consumption of the base case is com-pared against published data on the electrical energy consumptionof a residential house with similar dimensions in Beirut. Karaki et al.[24] reported that HVAC systems consume 50% of the total elec-trical energy consumption in residential houses in Lebanon at anapproximate consumption rate of 48 kWh/m2 [23], which is 10%different than the yearly energy consumption calculated by TRN-SYS [22]. Fig. 6 shows the monthly energy consumption calculatedby TRNSYS versus the results documented by Ghaddar and Bsat[27]. A maximum deviation of 10% is found between the publishedand the TRNSYS simulated monthly data.

4.3. Comparison against TRNSYS

In order to make sure that the proposed numerical model is

yielding accurate results for the zone conditions, the model isrun for the living zone on August 21st and November 21st. Thesimulation results of the ventilation airflow rates of the numericalmodel are used as input in the simulation of TRNSYS to compare the

ctrical consumption of HVAC system.

624 B. Yassine et al. / Energy and Bui

Fta

rSb

5

wscftpca

5

fitfsficHmc

ig. 7. Comparison between the numerical model and TRNSYS [22] software of theemperature and air mass flow rate in the living zone for the months of (a) August,nd (b) February.

esulting indoor air temperature of the numerical model and TRN-YS. Fig. 7(a) and (b) shows that the space temperature, predictedy the two models, is comparable with a difference of ±4%.

. Simulation results and discussion

As mentioned earlier, the base case apartment shown in Fig. 3,hich is divided into two zones (living and bedroom zones), is con-

idered as a test case for the proposed integrated space, thermalomfort, and ventilation control model. Simulations are carried outor a representative day, i.e. the 21st of each month of the year, forhe city of Beirut, Lebanon. In order to obtain accurate results inde-endent of the initial conditions, simulations are performed for 3onsecutive days, and the data for the third day are adopted in thenalysis described in the following section.

.1. Alternative cases

To attain the objective of identifying building materials and con-gurations which may optimize a building energy requirements,wo configurations, in addition to the base case, were examinedor each of the living and bedroom zones resulting in seventy-twoimulations (six simulations for each month). Several wall con-gurations of different building materials and wall thickness are

onsidered for the two zones, living and bedroom. In both zones,empcrete is used as an alternative for the conventional concreteix used for the masonry blocks. Hempcrete has lower thermal

onductivity and higher thermal capacitance compared with the

ldings 55 (2012) 618–630

conventional hollow concrete block, and a lower embodied energy[15].

The investigated configurations focus on selecting buildingmaterials that result in high insulation and capacitance in theliving room zone, and low insulation and capacitance in the bed-room zone. Two wall configurations were considered for the livingzone. Case 1 consists of 3 cm of straw (Cp = 2000 J/kg K) sand-wiched between 2 cm × 6 cm of Hempcrete (Cp = 1430 J/kg K). Case2 (massive case) consists of 5 cm of straw sandwiched between2 cm × 10 cm of Hempcrete. Two additional configurations (Cases3 and 4) were considered for the bedroom wall. Case 3 consists of10 cm of Hempcrete, and Case 4 is made of 10 cm of Hempcrete and5 cm of straw board placed on the internal side. The thermal proper-ties of straw were obtained from “the green building bible” [28] andthe thermal properties of hemp were obtained from Elfordy et al.[29]. Table 1 summarizes the various investigated configurations.

6. Results

Figs. 7–12 present the indoor air temperature for the differentwall configurations, for the months of (a) February, (b) April and(c) August, representative months for the winter, spring and sum-mer seasons. The fan power consumption is calculated based on theaffinity law for prime movers. The fan power is proportional to thecube of the flow rate ratio taken with respect to a reference. Duringthe month of February, the mechanical ventilation controller is setto the deactivation mode for both zones (living and bedroom) sincethe indoor air temperature is lower than the set winter temperatureand higher than the outdoor air temperature throughout the wholesimulation day resulting in a fan consumption of 0.96 kWh for allthe cases. For the living zone indoor temperature shown in Fig. 8(a)and (c), the base case wall configuration (Fig. 8(a)) characterizedby lower thermal capacitance showed an indoor air temperatureprofile that respond to the outdoor temperature variations muchmore than the wall configurations for Case 1 (Fig. 8(b)) and Case2 (massive wall, Fig. 8(c)). The thermal characteristics of the wallbase case resulted in having a total of 7 discomfort hours as com-pared to 6 discomfort hours for Case 1 and 3 discomfort hours forCase 2. The temperature inside the living zone of Case 2, shownin Fig. 8(c), tends to be constant during the non-occupancy period,and increases slightly during the occupied time. As for the bedroomzone, simulations during the month of February indicate that thewall with lowest insulation (Case 3) corresponds to the lowest airtemperature as shown in Fig. 9(b). This is due to a higher amountof heat loss between the indoor and the outdoor spaces. The wallwith the highest insulation (U = 0.98 W/m2 K, Cp,straw = 2000 J/kg K,Case 4) seems to have better conditions in the winter season sinceit has a higher indoor air temperature as can be seen in Fig. 9(c)and lower number of discomfort hours. Case 4 has five discomforthours occurring during the hour period 5 till 10 compared to six andseven discomfort hours for the base case and Case 3 respectively.

In the month of April (Fig. 10), night purging is applied in theliving zone, and since the temperature inside the space during theoccupied period is less than the set temperature, the controlleris set to the deactivation mode resulting in a fan consumption of3.32 kWh in the three cases. Both the base case and Case 1 showedan indoor living zone temperature higher than the outdoor air tem-perature with a general lower indoor air temperature profile forthe base case (Fig. 10(a) and (b)). As for the massive wall configu-ration (Case 2), the high capacitance of the wall results in havingan indoor air temperature that is lower than the outdoor during

the hour period of 8–15 and higher than the outdoor for the rest ofthe occupied period (see Fig. 10(c)). In the bedroom zone (Fig. 11),the control was also set to the deactivation mode during the occu-pancy time, resulting in a fan consumption of 0.96 kWh. The wall

B. Yassine et al. / Energy and Buildings 55 (2012) 618–630 625

Table 1Different examined scenarios of wall configuration of living and bedroom zones.

Zone Scenario Wall configuration U value (W/m2 K)

Both zone Base case 15 cm hollow concrete 2.36Living zone Case 1 3 cm of straw sandwiched between 2 cm × 6 cm of Hempcrete 1.37

Case 2 (massive case) 5 cm of straw sandwiched between 2 cm × 10 cm of Hempcrete 0.7Bedroom zone Case 3 10 cm of Hempcrete 3.1

Case 4 10 cm of Hempcrete + 5 cm of straw 0.986

Fig. 8. Temperature and mass flow rate in the living zone during February for (a)base case, (b) Case 1, and (c) Case 2 (massive case).

Fig. 9. Temperature and mass flow rate inside the bedroom zone during Februaryfor (a) base case, (b) Case 3, and (c) Case 4.

626 B. Yassine et al. / Energy and Buildings 55 (2012) 618–630

Fc

cptti(c

fl

ig. 10. Temperature and mass flow rate in the living zone during April for (a) basease, (b) Case 1, and (c) Case 2 (massive case).

ase of the highest insulation (Case 4) has the highest air tem-erature profile, and due to its highest thermal capacitance, theemperature inside the space takes more time to increase duringhe non-occupied period (see Fig. 11(c)). Case 3, which is character-zed by the lowest insulation, had the lowest temperature profile

see Fig. 11(b)). No discomfort hours were recorded in the differentases for both the living and bedroom zones.

In August, the controller modulated the outdoor ventilation airow to prevent the indoor air temperature from increasing beyond

Fig. 11. Temperature and mass flow rate inside the bedroom zone during April for(a) base case, (b) Case 3, and (c) Case 4.

the outdoor air temperature for both wall configuration, Cases 1and 2. Fig. 12(b) and (c) shows a lower indoor air temperature thanin the base case shown in Fig. 12(a). However, since comfort can-not be achieved at the indoor air temperature, the ceiling fan was

turned on to improve comfort conditions inside the space for theperiod hour extending from 12 to 20 and to limit the discomforthours to seven during the hour period of 13–20 in comparison to 12and 11 discomfort hours for the base case and Case 1 respectively.

B. Yassine et al. / Energy and Buildings 55 (2012) 618–630 627

Fb

IbttCahrw

ig. 12. Temperature and mass flow rate in the living zone during August for (a)ase case, (b) Case 1, and (c) Case 2 (massive case).

n the bedroom zone, Case 3 (Fig. 12(b)), which is characterizedy the highest heat transfer coefficient, had the lowest indoor airemperature at the early hours of the occupied period as comparedo the bedroom air temperature of the base case (Fig. 12(a)) andase 4 (Fig. 12(c)). Two hours of discomfort Thermal comfort was

ttained inside the space during occupancy with zero discomfortours for Case 3 against 2 h of discomfort for the two other bed-oom wall cases (base case and Case 4), moreover due to the lowerall insulation of Case 3, less ventilation air is needed to moderate

Fig. 13. Temperature and mass flow rate inside the bedroom zone during Augustfor (a) base case, (b) Case 3, and (c) Case 4.

the indoor zone temperature (5.70 kWh) as compared to the basecase (7.65 kWh) and (9.39 kWh) for wall Case 4 (Fig. 13).

The numerical model was run for the six simulations (three foreach of the living and bedroom zones) for the representative day ofeach month of the year. The results were multiplied by the num-ber of days in each month to determine the monthly number ofdiscomfort hours in each zone as shown in Fig. 14(a) and (b), and

the monthly fan consumption as shown in Fig. 15(a) and (b) for thedifferent scenarios.

As shown in Figs. 14 and 15, Case 2 (massive wall in the liv-ing zone) has the lowest operational energy and lowest number

628 B. Yassine et al. / Energy and Buildings 55 (2012) 618–630

ferent cases for (a) living zone and (b) bedroom zone.

ooccCChsoddob

sewda

Table 2Mechanical ventilation electrical consumption and number of yearly discomforthours for the different cases.

Configuration wall case Yearly electricalconsumption (kWh/m2)

Number of yearlydiscomfort hours

Base case (mechanicalheating and cooling)

53.53 Not applicable

Base case (mechanicalventilation)(living + bedroom)

31.80 2880

Living zone base case 14.32 2038Living zone Case 1 15.57 1794Living zone Case 2 9.02 975Bedroom zone base case 17.48 842

Fig. 14. Number of discomfort hours for the dif

f discomfort hours during the whole year as compared to thether living zone wall cases. As for the bedroom zone, the basease and Case 3 have approximately equal number of yearly dis-omfort hours (842 discomfort hour in the base case, and 815 inase 3); however, even though the high thermal wall insulation,ase 4, has the lowest number of discomfort hours (696 discomfortour) but it turns to be an in-effective solution for the bedroomince it has a 25.9% increase in the fan energy consumption. On thether hand, bedroom wall Case 3, compared to the base case, canecrease the fan energy consumption by 18.6% and decrease theiscomfort hours by 3.2%. However, the discomfort hours mainlyccur in the winter during hours when occupants are expected toe sleeping.

From Table 2, it is clear that the use of a mechanical ventilationystem for the base case can save around 40.5% of operational

nergy consumption of the conventional air conditioning systemhile having 2880 discomfort hours, 32.8% of discomfort hoursuring the year. The energy consumption can be reduced by 26.9%nd the discomfort hours can be reduced by 37.8% when using wall

Bedroom zone Case 3 14.22 815Bedroom zone Case 4 22.01 696

B. Yassine et al. / Energy and Buildings 55 (2012) 618–630 629

cases

cC

7

tmbfad

Biatita21tc

Fig. 15. Fan consumption for the different

onfiguration of Case 2 in the living zone and wall configuration ofase 3 in the bedroom zone.

. Conclusion

This paper integrates space and comfort and ventilation con-roller models into one model. Through simulation, the integrated

odel determines the amount of ventilation air that can possi-ly provide indoor thermal comfort for longer periods. This allowsor accurately measuring the energy performance of buildingsnd assessing the operational energy of various building envelopesigns.

In order to validate the simulation results, a typical house ineirut, Lebanon, is modeled in TRNSYS [22]. This required calibrat-

ng TRNSYS against published data, and then comparing the resultsgainst the proposed numerical model. The analysis confirmedhat the building design parameters affect the thermal comfortnside the space and fan consumption of the mechanical ventila-ion system. The simulation results revealed that a combination of

massive wall made of a 5 cm layer of straw sandwiched between cm × 10 cm of Hempcretein the living zone, and in addition to0 cm of Hempcrete in the bedroom zone, can reduce the opera-ional cost of the house by 26.9% and the discomfort hours by 37.8%ompared to the base case.

for (a) living zone and (b) bedroom zone.

Acknowledgment

The financial support of the Munib Masri Institute for Energyand Natural Resources at the American University of Beirut is highlyacknowledged.

References

[1] UNEP-SBCI, Buildings and Climate Change; Status, Challenge and Opportuni-ties, UNEP Publication, Paris, France, 2007.

[2] P. Kesselring, C.J. Winter, World energy scenarios: a two-kilowatt society,plausible future or illusion? in: Proceedings of the Energietage 94, Villigen,Switzerland November 10–12, 1994.

[3] A. Pfeiffer, M. Koschenz, A. Wokaum, Energy and building technology for the2000 W society potential of residential buildings in Switzerland Energy andBuilding 37 (2005) 1158–1174.

[4] T.F. Schulz, S. Kypreos, L. Barreto, A. Wokaum, Intermediate steps towards the2000 W society in Switzerland: an energy-economic scenario analysis, EnergyPolicy 36 (2008) 1303–1317.

[5] Le. Tran, C. Maalouf, T.H. Mai, E. Wurtz, F. Collet, Transient hydrothermalbehavior of a hemp concrete building envelope, Energy and Building 42 (2010)1797–1806.

[6] M. Kolokotroni, A. Aronis, Cooling-energy reduction in air-conditioned offices

by using night ventilation, Applied Energy 63 (4) (1999) 241–253.

[7] P. Blondeau, M. Spérandio, F. Allard, Night ventilation for building cooling insummer, Solar Energy 61 (5) (1997) 327–335.

[8] H. Breesch, Passive cooling in a low-energy office building, Solar Energy 79(2005) 682–696.

6 d Bui

[

[

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

30 B. Yassine et al. / Energy an

[9] G. Carrilho da Graca, Q. Chen, L.R. Glicksman, L.K. Norford, Simulation of winddriven ventilative cooling systems for an apartment building in Beijing andShanghai, Energy and Buildings 34 (2002) 1–11.

10] U. Eicker, ling strategies, summer comfort and energy performance of arehabilitated passive standard office building, Applied Energy 87 (6) (2010)2031–2039.

11] V. Geros, M. Santamouris, A. Tsangrasoulis, G. Gurracino, Experimental eval-uation of night ventilation phenomena, Energy and Buildings 29 (1999)141–154.

12] E. Shaviv, A. Yezioro, I.G. Capeluto, Thermal mass and night ventilation as pas-sive cooling design strategy, Renewable Energy 24 (3–4) (2001) 445–452.

13] B. Givoni, Indoor temperature reduction by passive cooling systems, SolarEnergy 85 (8) (2011) 1692–1726.

14] J. Zhou, G. Zhang, Y. Lin, Y. Li, Coupling of thermal mass and natural ventilationin buildings, Energy and Building 40 (2008) 979–986.

15] E. Awwad, B. Hamad, M. Mabsout, H. Khatib, Sustainable construction materialusing hemp fibers – preliminary study, in: Second International Conference onSustainable Construction Materials, June 28–June 3, 2010.

16] ASHRAE, ASHRAE Standard 55, Thermal Environmental Conditions for HumanOccupancy, ASHRAE, Atlanta, 1992.

17] ASHRAE, Climate, Comfort, and Natural Ventilation: A New Adaptive Comfort

Standard for ASHRAE Standard 55, University of California, Berkeley, 2001.

18] P.O. Fanger, Thermal Comfort, Danish Technical Press, Copenhagen, 1970.19] E. Kruger, Thermal monitoring and indoor temperature predictions in a pas-

sive solar building in an arid environment, Energy and Building 43 (2008)1792–1804.

[

[

ldings 55 (2012) 618–630

20] H. Zhang, E. Arens, C. Huizenga, T. Han, Thermal sensation and comfort modelsfor non-uniform and transient environments: part I: local sensation of individ-ual body parts, Building and Environments 45 (2) (2010) 380–388.

21] R. Dorf, R. Bishop, Modern Control Systems, 11th ed., Pearson InternationalEdition Pearson Education, 2008.

22] TRNSYS, Transient System Simulation Tool Software. Thermal Energy Sys-tem Specialists, LLC. Madison Wisconsin. http://www.tess-inc.com/, (accessed25.06.12).

23] A. Houri, S. Orfali, Residential energy consumption patterns: the case ofLebanon, International Journal of Energy Research (2005) 755–766.

24] S. Karaki, F. Chaaban, R. Chedid, T. Mezher, A. Harb, A. Abdulla, A. Hamzeh,A. Yahia, Electric energy access in Jordan, Lebanon and Syria, in: Proceedingsof International Conference on Energy Research and Development (ICERD-3),Kuwait, 2005.

25] Republic of Lebanon Ministry of Public Works and Transport. Energy Analysisand Economic Feasibility. UNDP/GEF, MPWT/DGU, 2005, p. 63.

26] United Nation Development Program, Lebanon Page, http://www.undp.org.lb/programme/pro-poor/poverty/povertyinlebanon/conditions97.cfm,(accessed 17.06.12).

27] N. Ghaddar, A. Bsat, Energy conservation of residential buildings in Beirut,International Journal of Energy Research 22 (1998) 523–546.

28] Hall Keith (Ed.), The Green Building Bible, vols. 1 and 2, 3rd ed., The GreenBuilding Press, 2006.

29] S. Elfordy, F. Lucas, F. Tancret, Mechanical and thermal properties of lime andhemp concrete (“hempcrete”) manufactured by a projection process, Construc-tion and Building Materials 22 (2008) 2116–2123.